14 lines
345 B
Python
14 lines
345 B
Python
from datasets import load_dataset
|
|
from pipeline import export_dataset
|
|
|
|
if __name__ == "__main__":
|
|
dataset = load_dataset(
|
|
"opencsg/chinese-cosmopedia",
|
|
)
|
|
export_dataset(
|
|
dataset=dataset["train"],
|
|
output_dir="./dataset",
|
|
output_prefix="chinese-cosmopedia-pretrain",
|
|
chunk_size=1_000_000,
|
|
)
|